cs.AI（2025-05-31）

📊 共 20 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗2) 支柱四：生成式动作 (Generative Motion) (1) 支柱一：机器人控制 (Robot Control) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 Alignment Revisited: Are Large Language Models Consistent in Stated and Revealed Preferences? 揭示大语言模型偏好偏差：一致性评估与可信赖性分析 large language model
2 FinBERT2: A Specialized Bidirectional Encoder for Bridging the Gap in Finance-Specific Deployment of Large Language Models FinBERT2：面向金融领域LLM部署的专用双向编码器，提升判别与检索性能 large language model
3 PMF-CEC: Phoneme-augmented Multimodal Fusion for Context-aware ASR Error Correction with Error-specific Selective Decoding 提出PMF-CEC，利用音素增强多模态融合，提升上下文感知ASR纠错中同音异形词的准确率。 multimodal
4 CMT-LLM: Contextual Multi-Talker ASR Utilizing Large Language Models CMT-LLM：融合上下文偏置的多说话人语音识别，利用大语言模型提升性能 large language model
5 ChartGen: Scaling Chart Understanding Via Code-Guided Synthetic Chart Generation ChartGen：通过代码引导的合成图表生成扩展图表理解能力 large language model multimodal ✅
6 Machine vs Machine: Using AI to Tackle Generative AI Threats in Assessment 提出一种基于机器对抗的AI评估框架，应对生成式AI在教育评估中的威胁 large language model multimodal
7 Position: Olfaction Standardization is Essential for the Advancement of Embodied Artificial Intelligence 呼吁AI领域重视嗅觉标准化，促进具身人工智能发展 multimodal
8 CodeSense: a Real-World Benchmark and Dataset for Code Semantic Reasoning CodeSense：提出一个真实世界代码语义推理的基准和数据集，用于评估和提升代码大模型在实际软件工程任务中的能力。 chain-of-thought ✅
9 RFCAudit: An LLM Agent for Functional Bug Detection in Network Protocols RFCAudit：利用LLM Agent检测网络协议中的功能性缺陷 large language model
10 Organizational Adaptation to Generative AI in Cybersecurity: A Systematic Review 网络安全组织通过调整框架和流程适应生成式AI，提升威胁建模和风险应对能力。 large language model
11 AgentAuditor: Human-Level Safety and Security Evaluation for LLM Agents AgentAuditor：提出一种基于记忆增强推理的LLM Agent安全评估框架，达到人类专家水平。 chain-of-thought ✅
12 MIRROR: Modular Internal Processing for Personalized Safety in LLM Dialogue MIRROR：模块化内部处理，提升LLM对话中的个性化安全 large language model
13 Wide Reflective Equilibrium in LLM Alignment: Bridging Moral Epistemology and AI Safety 利用广义反思均衡提升LLM对齐，增强伦理基础与动态可修正性 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
14 World Models for Cognitive Agents: Transforming Edge Intelligence in Future Networks 提出Wireless Dreamer，一种基于世界模型的强化学习框架，用于优化未来网络边缘智能。 reinforcement learning world model dreamer
15 A versatile foundation model for cine cardiac magnetic resonance image analysis tasks CineMA：用于电影心血管磁共振图像分析的多功能基础模型 masked autoencoder foundation model ✅
16 Dyna-Think: Synergizing Reasoning, Acting, and World Model Simulation in AI Agents Dyna-Think：融合推理、行动和世界模型模拟的AI Agent框架 imitation learning world model large language model
17 Reasoning Like an Economist: Post-Training on Economic Problems Induces Strategic Generalization in LLMs 提出Recon，通过经济学问题后训练提升LLM在多智能体系统中的策略泛化能力 reinforcement learning large language model ✅
18 BASIL: Best-Action Symbolic Interpretable Learning for Evolving Compact RL Policies 提出BASIL以解决可解释强化学习问题 reinforcement learning deep reinforcement learning

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
19 A "Wenlu" Brain System for Multimodal Cognition and Embodied Decision-Making: A Secure New Architecture for Deep Integration of Foundation Models and Domain Knowledge 提出“文lu”多模态认知与具身决策系统，安全融合知识与模型，赋能行业应用。 penetration foundation model multimodal

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
20 RiOSWorld: Benchmarking the Risk of Multimodal Computer-Use Agents RiOSWorld：评估多模态计算机使用Agent风险的基准测试 manipulation large language model multimodal ✅

⬅️ 返回 cs.AI 首页 · 🏠 返回主页